When in Doubt: Improving Classification Performance with Alternating Normalization
Machine Learning
2021-09-29 v1 Computation and Language
Computer Vision and Pattern Recognition
Abstract
We introduce Classification with Alternating Normalization (CAN), a non-parametric post-processing step for classification. CAN improves classification accuracy for challenging examples by re-adjusting their predicted class probability distribution using the predicted class distributions of high-confidence validation examples. CAN is easily applicable to any probabilistic classifier, with minimal computation overhead. We analyze the properties of CAN using simulated experiments, and empirically demonstrate its effectiveness across a diverse set of classification tasks.
Cite
@article{arxiv.2109.13449,
title = {When in Doubt: Improving Classification Performance with Alternating Normalization},
author = {Menglin Jia and Austin Reiter and Ser-Nam Lim and Yoav Artzi and Claire Cardie},
journal= {arXiv preprint arXiv:2109.13449},
year = {2021}
}
Comments
Findings of EMNLP 2021